Research map / 3D vision / Human pose estimation
Dual Grid Net: Hand Mesh Vertex Regression from Single Depth Maps
Chengde Wan, Thomas Probst, Luc Van Gool, Angela Yao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add lifting energy term (ablation) | Self-supervised Dual Grid Net · NYU hand pose (test) | mean joint error (mm): 14.50 -> 13.09 | Improved |
| add lifting energy term (ablation) | Self-supervised Dual Grid Net · NYU hand pose (test) | success frames at 30mm (%): 57 -> 64 | Improved |
| add active data augmentation by estimation (ablation) | Self-supervised Dual Grid Net · NYU hand pose (test) | mean joint error (mm): 14.52 -> 13.09 | Improved |
| add collision loss (ablation) | Self-supervised Dual Grid Net · NYU hand pose (test) | mean joint error (mm): 13.85 -> 13.09 | Improved |
| add ARAP loss (ablation) | Self-supervised Dual Grid Net · NYU hand pose (test) | mean joint error (mm): 14.06 -> 13.09 | Improved |
| add offset loss (ablation) | Self-supervised Dual Grid Net · NYU hand pose (test) | mean joint error (mm): 14.12 -> 13.09 | Improved |
| use multi-view training instead of single view (ablation) | Self-supervised Dual Grid Net · NYU hand pose (test) | mean joint error (mm): 16.96 -> 13.09 | Improved |
| train on larger training set instead of only the test samples (ablation) | Self-supervised Dual Grid Net · NYU hand pose (test) | mean joint error (mm): 14.50mm vs 13.09mm | Improved |
| apply kinematic refinement module to mesh vertices (ablation) | Synthetic-trained Dual Grid Net · Synthetic test (20K depth maps) | mean vertex error (mm): 14.75mm to 7.65mm | Improved |
| test synthetic-only trained network on real instead of synthetic data | Synthetic-trained Dual Grid Net · NYU hand pose | mean joint error (mm): 7.1mm to 23.21mm | Hurt |
| add keypoint supervision instead of self-supervision | Dual Grid Net · NYU hand pose (test) | mean joint error (mm): 13.1 -> 8.5 | Improved |
| evaluate on training set rather than unseen test set | Self-supervised Dual Grid Net · NYU hand pose | mean joint error (mm): 13.1mm to 11.9mm | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- Lie-X — Lie-X: Depth Image Based Articulated Object Pose Estimation, Tracking, and Action Recognition on Lie Groups (IJCV 2017) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- HandVoxNet: Deep Voxel-Based Network for 3D Hand Shape and Pose Estimation From a Single Depth Map (CVPR 2020 · similarity 0.90)
- 3D Hand Shape and Pose Estimation From a Single RGB Image (CVPR 2019 · similarity 0.90)
- HandVoxNet++: 3D Hand Shape and Pose Estimation Using Voxel-Based Neural Networks (TPAMI 2021 · similarity 0.90)
- Reconstructing Hands in 3D with Transformers (CVPR 2024 · similarity 0.88)
- Fine-Grained Multi-View Hand Reconstruction Using Inverse Rendering (AAAI 2024 · similarity 0.88)
- Learning Explicit Contact for Implicit Reconstruction of Hand-Held Objects from Monocular Images (AAAI 2024 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
